Table 2 Comparison of the results of different feature selection algorithms on the Framingham dataset.

From: Exploration and comparison of the effectiveness of swarm intelligence algorithm in early identification of cardiovascular disease

Algorithm

Population size

Average run time

Cohen Kappa value

Number of features

Avg

Max

Min

Std

WOA

10

27.0 s

0.62884

0.64141

0.61692

0.00870

10

25

59.8 s

0.63430

0.64177

0.62548

0.00496

11

50

122.2 s

0.63824

0.64301

0.63356

0.00297

10

CSA

10

89.0 s

0.59783

0.78417

0.50000

0.08501

11

25

208.3 s

0.57390

0.76668

0.49476

0.07906

13

50

389.7 s

0.57142

0.68143

0.50000

0.05389

10

FPA

10

581.4 s

0.62097

0.78954

0.49870

0.07814

9

25

1437.5 s

0.62898

0.77934

0.52452

0.06370

10

50

1739.2 s

0.58342

0.69343

0.51200

0.04536

10

HHO

10

458.7 s

0.61950

0.72280

0.48208

0.06018

12

25

1063.9 s

0.62097

0.68280

0.58782

0.02375

11

50

2191.2 s

0.60763

0.68280

0.56929

0.02838

12

PSO

10

1621.5 s

0.46231

0.61200

0.38576

0.05656

12

25

1824.9 s

0.46474

0.59000

0.38726

0.05069

11

50

2022.3 s

0.52717

0.63167

0.30000

0.08292

12

GA

10

97.1 s

0.43531

0.60000

0.28576

0.09252

10

25

628 s

0.44474

0.60000

0.39736

0.07841

12

50

1448.9 s

0.53617

0.63153

0.29999

0.09145

11

Back to article page